Cement & Concrete Composites 61 (2015) 3643

Contents lists available at ScienceDirect

Cement & Concrete Composites

journal homepage: www.elsevier.com/locate/cemconcomp

Performance evaluation of structural concrete using controlled quality

coarse and ne recycled concrete aggregate
Abdurrahmaan Lotfy , Mahmoud Al-Fayez 1
Lafarge Canada Inc., 6509 Airport Road, Mississauga, ON L4V 1S7, Canada

a r t i c l e i n f o a b s t r a c t

Article history: This paper presents the fresh, mechanical, and durability performance, of a structural concrete mix clas-
Received 20 August 2014 sied as C-1, by the Canadian Standards Association (CSA) made with controlled quality Recycled
Received in revised form 24 February 2015 Concrete Aggregate (RCA). Five mixes with water-to-cementing material (w/cm) ratio of 0.40 were pro-
Accepted 27 February 2015
duced with various RCA contents and tested against two 0% RCA control mixes made with General Use
Available online 10 April 2015
(GU) cement, and General Use Limestone cement (GUL). The RCA contents in the mixes were 10%, 20%,
and 30% by coarse aggregate volume replacement, as well as 10% and 20% ne and coarse (granular)
Keywords:
aggregate volume replacement. All evaluated mixes met the specications from the CSA for fresh,
Sustainability
Recycled concrete aggregates
mechanical, and durability properties. The coarse RCA mixes performed better than the granular RCA
Structural concrete mixes in terms of exural and splitting tensile strengths, linear drying shrinkage, water sorptivity, and
Fresh properties rapid chloride-ion permeability, where the test results were signicantly affected by the ultra nes pre-
Mechanical properties sent in the granular RCA.
Durability 2015 Elsevier Ltd. All rights reserved.

1. Introduction
The implementation of environmentally friendly construction
methods is becoming more widespread as awareness of the need
for sustainability in design grows. An ever-increasing rate of
construction is occurring with new buildings and thus demanding
concrete, as a durable and widely used product in these structures.
It is therefore necessary to nd ways to make concrete a more
environmentally conscious material. Crushing concrete from
demolished or deconstructed structures provides Recycled
Concrete Aggregates (RCA), which can be used as a partial substi-
tute of virgin aggregates in new concrete production. Not only will
recycled concrete aggregates signicantly decrease the demand for
quarried stone, thus decreasing transportation costs and emissions,
but these aggregates will also, in turn, be diverted from landlls. In
the United States alone, an estimated 200300 million tons of
construction and demolition waste is created each year [1]. A lot
of this waste could be re-used by crushing the concrete waste
and using it as RCA.
RCA has already been widely used as granular material in
pavement design. However, extensive research has been, and is
currently being, conducted in many countries on the performance
Corresponding author. Tel.: +1 905 738 7070.
E-mail address: abdurahman.lotfy@lafarge.com (A. Lotfy).
1
Tel.: +1 905 738 7070.
of RCA in structural applications [2]. The use of RCA in high perfor-
mance concrete is not widely accepted as of yet, primarily due to a
few reductions in mechanical properties as well as durability
which may be observed. It has been found that cement paste in
RCA contributes to a lowered relative density and higher water
absorption than virgin aggregates, while higher shrinkage and
creep strains were also observed. However, coarse RCA was shown
to provide similar compressive and exural strength results to vir-
gin aggregates [3]. Tighe and Butler [4] suggest that a good RCA
should meet certain criteria in order to be suitable for use in rein-
forced concrete. The criteria include: an aggregate relative density
of 2.3 or higher (to contribute to bond strength), a maximum mor-
tar content of 50%, and a maximum absorption of 3%.
The research presented in this paper is based on experimental
results from testing 35 MPa structural concrete with RCA. Two
different gradations of RCA were used in this study; one was
200 mm RCA (which includes nes) and the other was
207 mm RCA, which was used as a substitution for virgin coarse
aggregate. These two gradations of RCA were used at various
replacement levels to achieve a good understanding of the con-
cretes fresh, mechanical, and durability performance with varying
quantities of RCA.
The objective of this study is to analyze the potential of
Recycled Aggregate Concrete (RAC) in structural applications. The
use of RAC has been limited to low-risk applications such as side-
walks, residential wall mixes, backll, and as a granular material

http://dx.doi.org/10.1016/j.cemconcomp.2015.02.009
0958-9465/ 2015 Elsevier Ltd. All rights reserved.
 A. Lotfy, M. Al-Fayez / Cement & Concrete Composites 61 (2015) 3643
for pavements as designers are hesitant to include RCA into struc-
tural, reinforced concrete mixes. The potential of RAC to be used in
structural high-risk applications was studied by comparing the
performance of the RAC trial mixes to a control. This study used
a commercially produced RCA which is subjected to a high level
of quality control thus permitting the use of RCA into structural
concrete even further.
2. Research signicance
Incorporating RCA in structural concrete applications poses a
higher risk factor than the current applications of recycled aggre-
gate materials (i.e. granular base or subbase in roads, granular
backll for retaining walls, etc.). It is therefore important to under-
stand the behavior of a C-1 concrete mix made with coarse and ne
aggregates which have been partially substituted with RCA. In
many cities, old buildings are being demolished and new ones
are being constructed. From an environmental and cost savings
standpoint, it is very logical to take waste from one site and re-
use it in a new project. Literature discusses various drawbacks of
using RCA in concrete, such as increased water demand and
decreased strength. By demonstrating the equivalent performance
exhibited through replacement of up to 30% RCA (by volume of
respective aggregate) into a 35 MPa C-1 structural concrete mix
with a water-to-cementing material (w/cm) ratio of 0.40, this
study aims to identify the further potential of RCA and reduce
the stigma of its commercial feasibility and performance.
3. Experimental program
To properly assess the performance of the controlled quality
RAC, a full spectrum of tests were conducted on the fresh, mechan-
ical, and durability properties. The following tests were performed
on all mixes; slump, measured air, density, compressive strength,
exural strength, split tensile strength, linear drying shrinkage,
Rapid Chloride Permeability (RCP), resistance to freezing and thaw-
ing, sorptivity, as well as microscopic analysis. Since the applica-
tion for this mix is a structural reinforced concrete, a vigorous
testing program was performed to test every aspect of the con-
cretes properties. Five mixes with varying replacement levels of
RCA were tested and compared against the control. The type of
RCA was a commercially produced controlled quality aggregate,
subjected to a very high level of quality control with a deleterious
material content much less than 1%.
The concrete mixtures compositions and identication are
provided in Table 1. The mix design chosen was a conventional
concrete mix with a desired compressive strength of 35 MPa at
28 days. This mix design meets the exposure class of C-1 according
to the Canadian Standards Association (CSA) A23.1; reinforced con-
crete exposed to freezing and thawing, as well as chlorides. Air
entraining admixture, as well as mid-range water reducing admix-
ture, was used in all mixes. The CSA A23.1 specications on the mix
state: the w/cm ratio must be limited to 0.4; it must reach a min-
imum of 35 MPa compressive strength at 28 days, as well as the
RCP test must pass less than 1500 C at 56 days. The fresh property
specications according to A23.1 state: the mix must have an air
content of 58%, as well as a slump of 80 30 mm. The mix design
uses Type 10 GU cement as well as a GUL cement equivalent to
ASTM Type 1. It also incorporates 35% Ground Granulated Blast
Furnace slag (GGBF slag or slag), by mass of the total cementing
material content.
The testing program consisted of comparing the results of seven
mixes with varying percentage replacement levels of the RCA. Two
variations of recycled concrete aggregate were used, a 207 mm
coarse gradation, and a 200 mm granular full gradation. These
37
gradations were tested at different replacement levels and were
compared to the control mix. RCA was replaced volumetrically with
the virgin aggregate. Coarse RCA replaced only the coarse virgin
aggregate at levels of 10%, 20%, and 30%, while the granular full gra-
dation was added to replace the same proportion of coarse and ne
aggregates in the control mix at levels of 10% and 20%. The seven
mixes used in this study are: (GU) control mix with Type GU
cement, (GUL) control mix with Type GUL cement, (C10) 10% coarse
RCA, (C20) 20% coarse RCA, (C30) 30% coarse RCA, (G10) 10% gran-
ular RCA, and (G20) 20% granular RCA, as listed in Table 1. With
the exception of the Type GU benchmark, all mixes incorporated
the use of Contempra Portland Limestone Cement (PLC Type
GUL), a product known industry-wide to support sustainable con-
struction practices through its reduced Carbon footprint.
3.1. Materials
Other than the GU cement control mix, all the mixes used GUL
cement equivalent to ASTM Type 1, and GGBF slag. The chemical
composition and physical characteristics of these cementing mate-
rials are presented below in Table 2. Virgin aggregates used in the
mixes were 19 mm nominal size, crushed limestone and concrete
sand with specic gravities (SSD) of: 2.57 and 2.68, respectively.
Coarse RCA and granular RCA were both derived from commercial
production and tested for performance characteristics as discussed
in Section 3.1.1, and as shown in Table 3. Blends incorporating
coarse RCA were tested for physical properties, as can be seen in
Table 3. All aggregate gradations are shown in Fig. 1.
3.1.1. Controlled quality recycled concrete aggregates
Recycled concrete aggregates can be of great quality, and
meet all the specications as an aggregate in concrete [5]. What
inhibits its use is its variability, and the high risk associated with
decient structures. The type of RCA used in this study is a con-
trolled quality RCA. The RCA is processed on a commercial scale
to a high and consistent quality. RCA may usually contain many
contaminants such as: metal, asphalt, organics, glass, ceramics,
plaster, drywall (gypsum) board, roong materials, sealants, and
wood. The total deleterious material content in the RCA used in
this study was less than 1%, which is less than the CSA standards
limit; a total of 2% deleterious materials content.
The production process of the RCA used in this study limits the
types of concrete being selected for crushing to: returned concrete,
concrete wash out material, and waste concrete rubble and blocks.
Each load of recycled material is continuously screened for any
deleterious materials and unsuitable materials. Gradation, and
testing for asphalt coated particles, is performed every 1000 metric
tons, while other tests are frequently performed every 25,000 tons
during production. This type of RCA has a controlled level of quality
and is thus able to produce repeatable, adequate results.
3.2. Test methodology
3.2.1. Fresh properties
Each mix was tested for slump retention up to 45 min, as well
as air content. The slump was conducted after 15 min of mixing,
while the slump retention was performed at 45 min after mixing
to represent actual on-site conditions. Slump testing was per-
formed according to ASTM C143, while the pressure method was
used for the measured air content testing according to ASTM C231.
3.2.2. Mechanical properties
The compressive, exural, and split-tensile strengths were
determined for each mix. These three mechanical properties were
meant to give a better understanding of the physical interaction of
the RCA in the concrete mix. Compressive strength was tested on
38 A. Lotfy, M. Al-Fayez / Cement & Concrete Composites 61 (2015) 3643

Table 1
Mix design composition.

Mix w/ Cement Slag Virgin coarse Virgin ne Coarse RCA Granular RCA Air entraining Mid-range water
ID c type replacement aggregate info aggregate info replacement replacement (%) admixture (l/m3) reducing admix. (l/m3)
(%) (%)
GU 0.4 GU 35 CSA Certied 20 mm CSA Certied 0 0 0.20 0.34
GUL 0.4 GUL 35 Crushed Limestone Concrete Sand 0 0 0.20 0.34
C10 0.4 GUL 35 10 0 0.20 0.34
C20 0.4 GUL 35 20 0 0.20 0.34
C30 0.4 GUL 35 30 0 0.20 0.34
G10 0.4 GUL 35 0 10 0.20 0.34
G20 0.4 GUL 35 0 20 0.20 0.34

Table 2 critical, for the same reason, thus sorptivity tests were also per-
Cementitious compositions of the cements used. formed for each mix according to ASTM C1585. Linear drying
Cementitious properties shrinkage tests were performed according to ASTM C157 on con-
Chemical GU cement GUL cement Slag
crete prisms with a cross section of 75 mm  75 mm (300  300 ).
SiO2 (%) 19.6 18 38.4
Al2O3 (%) 4.9 4.6 10.64 4. Results and discussion
Fe2O3 (%) 3.1 2.9 0.79
TiO2 (%) 0.71 4.1. Fresh properties
CaO (%) 61.4 60.1 34.2
MgO (%) 3 2.5 6.94
SO3 (%) 3.6 4.1 0.2 In order to best mimic conditions in the eld, slump and air
Alkalis as Na2O (%) 0.7 0.62 0.16 retention were measured at 15 min and again at 45 min to account
LOI (%) 2.3 5.5 3.09 for travel time from the plant to the construction site. Since a w/cm
Blaine (cm2/g) 3870 4880 4500
ratio of 0.4 was adhered to, it was necessary to add a High Range
+45 lm (%) 3 0.4 1
Density (g/cm3) 3.15 3.05 2.93 Water Reducer (HRWR) to each mix to achieve workable results.
The slumps of all seven mixes, including the two benchmark mixes,
were in the range of 105 10 mm, and all nal slump values were
concrete cylinders with a diameter of 100 mm according to ASTM
in the range of 75 5 mm. These results were consistent with those
C39, while exural strength testing was performed on rectangular
of other studies in which the nal slump of the RCA mix with
beams with dimensions as shown in Table 4. The exural strength
HRWR was found to be in the range of 80 10 mm [6]. The air con-
was tested using 3-point loading according to ASTM C78. Split ten-
tent of all seven mixes was in accordance with CSA standards
sile strength was determined using concrete cylinders with a
which ranged from 5% to 7.2%.
diameter of 100 mm, as outlined in ASTM C496. All mechanical
A common challenge in obtaining suitable fresh properties in
properties were determined for 7 and 28 days.
RAC comes from the paste attached to the stones which increases
porosity and water demand. This residual paste is relatively weak,
3.2.3. Durability properties and when it is handled, transported, or mixed, it experiences sig-
Structural concrete in a Canadian climate can be exposed to sev- nicant abrasion. The abraded material is ner and increases the
ere conditions, which included rapid freezing and thawing cycles, water demand of the mix. For this reason, all RCAs were used in
as well as daily chloride exposure. It is thus very important that a Saturated Surface Dry (SSD) condition, as suggested by West
the durability properties are not affected by the addition of RCA. et al. [7]. Fig. 2 summarizes the fresh property results.
The seven mixes were tested for their freezing and thawing resis-
tance, according to ASTM C666, for 200 cycles. Since this C-1 class 4.2. Mechanical properties
mix is steel reinforced according to CSA A23.1, the salt penetration
to the steel reinforcement is critical to the durability of the struc- 4.2.1. Compressive strength
ture. RCP testing was performed at 28 and 56 days, according to The minimum requirement for a C-1 Class mix for compressive
ASTM C1202. The absorption of water into the concrete is also strength is 35 MPa at 28 days according to CSA A23.1. Fig. 3 shows

Table 3
Aggregate properties for: virgin, RCA aggregates as well as the coarse RCA blends.

CSA A23.2-2A test Spec Concrete 19 mm Max virgin RCA granular RCA coarse Blend (RCA coarse
sand limestone (200 mm) (207 mm) aggregate:Virgin
limestone aggregate)
10:90 20:80 30:70
% Micro-Deval abrasion loss 14/17 9.6 7.3 21.1 20.8 9.1 10.2 11
max
% Flat and elongated 20 max 10 1 2 5 3 3
Accelerated mortar bar 0.150 max 0.036 0.051 0.045 0.043 0.048
Magnesium sulfate 12 max 2.8 2.8 3.6 0.6 0.5 1.1
soundness
% Wash loss (80 lm) 2 max 0.9 0.8 4 2.8 0.5 0.6 0.7
Low density particles 0.5/1 1.24 0.01 0.03 0.39
% Absorption 2 max 0.6 0.89 11.57 5.4 1.2 1.54 1.9
Relative density bulk 2.665 2.732 2.047 2.362 2.692 2.661 2.621
Relative density SSD 2.681 2.754 2.284 2.489 2.724 2.72 2.67
Relative density apparent 2.707 2.793 2.683 2.707 2.782 2.775 2.758
 A. Lotfy, M. Al-Fayez / Cement & Concrete Composites 61 (2015) 3643 39

120 (compressive and exural strength) are equivalent to the control.

If normal concrete is used, then incorporating RCA is a very viable
100
Percentage Passing (%)

option in structural applications, because there will be little to no

80
60
40
20
0 slight reduction in strength is usually attributed to the low density,
0.01 0.1
Parcle Size (mm)
Coarse RCA (20 -7 mm)
19 mm Limestone
Fig. 1. Aggregate gradations.
constant compressive strength results for 7 and 28 days for the
seven mixes. This trend is consistent with previous studies in
which the RCA content did not alter the compressive strength,
given that the concrete was a normal strength type typically hav-
ing a w/cm ratio of 0.40.6 [8]. Otsuki [8] found that for normal
strength concrete, the type of aggregate made little difference to
the compressive strength, because the Interfacial Transition Zone
(ITZ) was the controlling phase in the concrete (i.e. the weakest
phase). This explains why an aggregate with higher porosity and
lower strength can produce concrete of similar strength, if the con-
crete is a normal strength concrete.
Some researchers found that the compressive strength
decreased as the replacement level increased [911], and that the
coarse RCA mixes produced higher strength than ne RCA concrete
mixes [12,13]. This was not found in this study and it could be due
to the variation in the type of RCA. The type of RCA used in this
study was a controlled quality RCA produced at a commercial scale.
This RCA, as explained in Section 3.1.1, goes through vigorous qual-
ity control and is maintained to be consistently at a high quality. As
the results show, there is a clear distinction when this type of
aggregate is used compared to conventional RCA. The RCA per-
formed equally to the virgin aggregate up to 30% coarse replace-
ment, and 20% granular replacement.
4.2.2. Flexural strength
Fig. 4 shows the exural strength results for 7 and 28 days for
the seven mixes. The exural strength also showed the same trend
as the compressive strength, but at a lower degree. Virtually no dif-
ference was observed between exural strength as the replace-
ment level of RCA increased. There is a slight pattern in which
the exural strength reduces as the RCA content increases, but this
decline is too marginal to consider. This property promotes the use
of RCA in structural concrete because the two major properties
1 10
Granular RCA (20 -0 mm)
Concrete Sand
reduction in performance.
4.2.3. Splitting tensile strength
The splitting tensile strength of all seven mixes is shown in
Fig. 5 for 7 and 28 days. A slight decline in the splitting tensile
strength is noticed as the replacement of RCA is increased. The
100
high porosity, and low strength of the RCA particles, but may not
always be the case. For normal strength concrete, it was deter-
mined that the controlling phase of the concrete was the ITZ [8].
This phenomenon explains why the compressive and exural
strengths showed no difference in strength as the virgin aggregate
was replaced with RCA, but this is not consistent with the splitting
tensile strength. This could be due to other factors that may reduce
the splitting tensile capability of the concrete, such as angularity of
the particles. According to Sagoe-Crentsil [14], the processing of
RCA produces more rounded particles than those produced in the
laboratory. Since the RCA in this study was produced on a commer-
cial scale, the continuous processing, handling, and transportation,
abrades the RCA and produces smoother particle shapes. As such,
RCA particles resemble a gravel particle shape rather than crushed
limestone. This reduced angularity of the RCA particles, compared
to the virgin limestone aggregate, reduces the restraint given to the
concrete matrix, and in turn, results in lower split tensile strengths.
The splitting tensile strength is more controlled by the paste
strength as well as the paste-aggregate bond, and a better bond
is provided with a more angular particle shape [15].
The coarse RCA outperformed the granular RCA in terms of split
tensile strength. Padmini et al. [16] found the same observation in
which the strength increased as the maximum size of RCA
increased. This is due to the reduction in nes when the coarse
RCA aggregate is added, compared to the granular RCA, as well as
the quality of those nes replaced. Even as the replacement of
coarse RCA reaches 30%, the split tensile strength is still higher
than the 10% Granular RCA. This makes coarse RCA a more prefer-
able candidate when replacing virgin aggregate.
According to Meddah et al. [17], strength of normal strength
concrete increases as the maximum size of the aggregate increases.
Also, according to Otsuki et al. [8], the controlling factor for com-
pressive strength, in normal strength concrete, is the ITZ between
the paste and the aggregates. For this reason, as the aggregate
120 9
8
100
7
80 6
Air Content (%)
Slump (mm)

Table 4 5
60
Mechanical and durability properties testing protocol.
4
Mechanical and durability testing protocol
40 3
Parameter Specimen Specimens Total # of Dimensions
type per mix specimens (mm) 2
20
Compressive Cylinder 18 90 100  200 1
strength
Flexural Beam 2 10 540  155  155 0 0
strength GU GUL C10 C20 C30 G10 G20
Drying Prism 3 12 285  75  75 15 Minutes Air Content (%) 15 min
shrinkage 45 Minutes Air Content (%) 45 min
Rapid chloride Cylinder 4 20 100  200 Slump Upper Limit Air Content Upper Limit
permeability
Slump Lower Limit Air Content Lower Limit
Freezethaw Prism 3 12 285  75  75
resistance
Fig. 2. Slump retention, and air content of mixes at 15 and 45 min.
40 A. Lotfy, M. Al-Fayez / Cement & Concrete Composites 61 (2015) 3643

50 mixes, produces a weaker strength concrete compared to the

45
coarse RCA mixes.

40
4.3. Durability properties
Compressive Strength (MPa)

35
30
25
20
15
10
5
0 outperforming the control mixture due to the air-entrained mortar
GU GUL C10
7 Days
Fig. 3. Compressive strength results at 7 and 28 days.
becomes coarser, surface area is decreased, and less ITZ area is pre-
sent. Since the ITZ is the controlling factor, in other words, the
weakest component of the concrete matrix, having less ITZ will
produce higher strength compared to a concrete with evidently
more ITZ. This is why Meddah et al. [17] found that increasing
maximum aggregate size, in turn increases strength. Hence ner
gradations produced weaker strength in normal strength concrete.
This principle applies in the trial mixes which displayed inferior
performance for the granular RCA relative to coarse RCA.
In addition to the increased presence of nes in the granular
RCA mixes, the reduction in strength is also attributed to the
reduced quality of the nes replaced. Coarse RCA contains rem-
nants of virgin aggregate and residual paste. In general, the resid-
ual paste on the RCA particle is weak [18] (washout concrete;
weaker initial concrete), thus the coarser particles have a higher
strength because the two components (residual paste and aggre-
gate) experience a better bond. West et al. [4] suggested that the
maximum mortar content of RCA should be less than 50%. Finer
RCA particles are more likely to be composed of residual paste.
As the RCA is processed and handled, it is very easy for the nes
to wear off (abrade), as residual paste has a reduced hardness.
This trend is observed in Fig. 5. Even with higher coarse RCA
replacement, splitting tensile strength is still higher than a smaller
replacement level of granular RCA. The combination of the increase
in nes, as well as the quality of the nes, in the granular RCA
4.3.1. Resistance to freezing and thawing
The performance of each mix was tested after approximately
300 cycles of freezing and thawing. Durability factor results are
shown in Table 5.
All specimens displayed adequate durability throughout the
cycles, and the C30 mix (with 30% coarse RCA) outperformed even
the two benchmark mixes. Similar performance has been observed
in other research where coarse RCA was used as a replacement to
natural coarse aggregate and improved its freeze thaw resistance,
C20 C30 G10 G20 in the RCA [19,20]. However, a signicant decline in performance
28 Days CSA Standard and durability was observed after only 90 cycles for the concrete
for 28 Days mixes with RCA in which the attached mortar was not air
entrained [19]. This suggests that in climates which undergo freez-
ing and thawing, the RCA in a C-1 mix should contain residual mor-
tar which has been air-entrained. This is a challenge that concrete
producers will face due to the uncertainty of the parent concrete
material, but may be mitigated with adequate quality control of
air admixture dosages during production. Also, in a country such
as Canada, a signicant amount of the concrete placed is typically
air entrained. Even if it is hard to distinguish between the sources
of the RCA, it will be likely that a signicant portion of the RCA
would have come from air-entrained concrete. Overall, all of the
mixes exhibited adequate performance and resistance to freezing
and thawing cycles, as shown in Table 5.
4.3.2. Linear drying shrinkage
Fig. 6 displays the drying shrinkage results for all seven trial
mixes. The drying shrinkage experienced in all the mixes is rela-
tively similar with all the results at 28 days within 0.015% of the
control mix. It was expected that linear drying shrinkage would
be higher for the RCA mixes, as seen in previous unpublished trials
by the authors conducted in controlled laboratory trials, given the
higher porosity of the attached mortar phase, and thus overall
higher paste content. However, the difference was not observed
in this study. The coarse RCA displayed equivalent performance
to the control mixture, while the granular RCA mixes resulted in
an overall higher drying shrinkage. This trend is similar to previous
studies with RCA [21], and stated that the inclusion of HRWR sig-
nicantly reduced drying shrinkage in RAC. This trend is similar to
the trends found in the mechanical properties testing, and can be

5
9
4.5
8
4
Split Tensile Strength (MPa)
Modulus of Rupture (MPa)

7 3.5

6 3

5 2.5

4 2

3 1.5

2 1

1 0.5

0 0
GU GUL C10 C20 C30 G10 G20 GU GUL C10 C20 C30 G10 G20
7 Day 28 Day 7 Day 28 Day

Fig. 4. Flexural strength results at 7 and 28 days. Fig. 5. Splitting tensile results at 7 and 28 days.
 A. Lotfy, M. Al-Fayez / Cement & Concrete Composites 61 (2015) 3643 41

Table 5 1800
Resistance of the specimens to freezing and thawing cycles.
1600

Charge Passed (Coloumbs)

No. of cycles GU GUL C10 C20 C30 G10 G20 1400
0 100 100 100 100 100 100 100
1200
44 99 97 98 98 101 99 97
91 99 98 98 100 101 100 98 1000
145 99 98 98 97 101 101 100
800
195 100 98 99 99 101 101 101
248 100 98 100 97 99 101 101 600
300 100 98 99 97 99 100 101
400

200

0
attributed to the increased nes, as well as the reduced quality of
GU GUL C10 C20 C30 G10 G20
the ne RCA, compared to virgin concrete aggregate. 28 Days 56 Days CSA Standard

4.3.3. Rapid chloride permeability Fig. 7. Rapid chloride permeability of specimens at 28 and 56 days.
All of the mixes in this study exhibited relatively low chloride
ion penetrability at 28 days, as per ASTM C1202, in the range of
0.016
10002000 C. CSA species that a maximum passing charge of
1500 C must be met by 56 days, and as is shown in Fig. 7, one of 0.014

Sorpvity (mms -0.5)

the mixes had already satised that requirement at 28 days 0.012
(C10). As expected, increasing the content of RCA in a mix also 0.01
increased the chloride ingression; however, no signicant increase
0.008
was observed between mixes. Research has shown that RAC with-
0.006
out Supplementary Cementing Materials (SCMs) tends to have
higher permeability as compared to RAC made with SCMs [22]. 0.004

Berndt [23] found that incorporating slag in a concrete mix signif- 0.002
icantly decreased the permeability, as the SCM particles take up 0
pore space and also help counteract the increase of permeability GU GUL C10 C20 C30 G10 G20
from the addition of RCA. In addition, a lower w/cm ratio decreases INITIAL SECONDARY

the permeability of concrete, which is benecial in a concrete with

Fig. 8. Initial and secondary gradient of sorptivity of each specimen.
RCA, in which the adhered/residual paste will tend to increase the
permeability.

wet conditions [25]. The sorptivity results in this study tended to

4.3.4. Sorptivity
increase as the percentage of replacement RCA increased. The
Two cylindrical discs for each mix were placed into 2 1 mm of
C30 mix was shown to have the highest sorptivity value. Other
water on a support device (as per ASTM C1585) and the rate at
studies have shown that a replacement of coarse aggregate with
which the water was absorbed into the concrete was measured
coarse RCA up to 25% did not increase or decrease the sorptivity
and plotted in Fig. 8 for initial and secondary gradient. The discs
of the concrete, and these concretes had a signicantly lower sorp-
were 50 mm deep and were cut from 100 mm by 200 mm concrete
tivity when the mix was designed with superplasticizers and a
cylinders. Initially, 25 mm from the top and bottom were cut off;
lower w/cm ratio [26]. These results are not consistent with those
the remaining 150 mm is cut into three discs of 50 mm depth each.
obtained in this study. However, a sudden increase was noticed
Only the middle 50 mm disc is used for this test, and the remaining
when the percentage of replacement of coarse RCA increased from
two discs are discarded. Testing specimens for sorptivity gives an
20% to 30% for secondary sorptivity, which was also observed by
indication of a concretes durability and can be related to the
Medina et al., when the replacement of RCA was increased from
degree and quality of curing. The quality of the concrete near the
25% to 50% [26]. The initial sorptivity of granular mixes for both
surface is determined by curing and this affects the rate at which
10% and 20% replacement were found to be marginally higher than
the sample will absorb water [24]. The test is performed in such
their coarse RCA counterparts. It is believed that this was observed
a way that the absorption of water is in one direction.
due to the overall increased mortar content, exhibited by the gran-
A higher sorptivity value is indicative of a highly interconnected
ular mixes, relative to the coarse mixes with the same volumetric
pore network that lls with water when the surface is exposed to
replacement level. However, it is to be noted that a larger volume
of aggregate was replaced due to the presence of the ultra ne
0 materials in the granular RCA mixes. The increase in the overall
-0.01 higher mortar content in the granular RCA mixes is reected in
-0.02 the secondary sorptivity results. It is believed that the pore struc-
Length Change (%)

-0.03 ture, being made up of a higher volume of cementitious material in

-0.04 the granular RCA mixes, would have a higher afnity for water. It is
-0.05 also found that, with increasing RCA replacement level, more
-0.06 reduction from the secondary sorptivity results is observed relative
-0.07 to the initial results. This is also due to the fact that the RCAs pore
-0.08 structure has a higher capability to absorb additional water over
-0.09 time than virgin aggregates.
0 10 20 30 40 50 60 70 80 90
Research suggests that lengthening the curing time of the con-
Time (days)
crete can help reduce sorptivity, by allowing more hydration to
GU GUL C10 C20 C30 G10 G20
occur, and thus decreasing the porosity of the concrete [25]. This
Fig. 6. Linear drying shrinkage of each specimen over 77 days. measure would help in concretes which have the challenge of
42 A. Lotfy, M. Al-Fayez / Cement & Concrete Composites 61 (2015) 3643

coarse aggregate. As can be seen, it exhibits a very dense and

homogenous microstructure, making it ideal for use in this applica-
tion. The presence of sand, stone, and cement paste, indicates that
this particle is from a reclaimed concrete, and not washout con-
crete, which will tend to be more segregated. Using RCA of con-
trolled high quality produces concrete of higher and more
consistent quality.

5. Conclusions

The results from the study are summarized below:

1. Through Controlled-quality industrial grade production, RCA is

Fig. 9. Control GUL mix.
Fig. 10. RCA mix with 30% coarse replacement.
increased porosity due to inclusion of RCA. Inclusion of SCMs and
superplasticizers has been suggested to help lower the sorptivity
in RAC, as they help to lower the absorption of water into the con-
crete [26]. It was also noted that the sorptivity decreased with a
lower w/cm ratio. Houehanou et al. [25] tested a mix with GU
cement, a 35% slag replacement, and a w/cm ratio of 0.42, and com-
parable results were found between the two studies.
4.4. Microscopic analysis
Figs. 9 and 10 show a polished cross section of the GUL control
mix, as well as magnied pictures of the RCA particles, respec-
tively. There is a uniform dispersion of aggregates across the cross
section for both the control mix (Fig. 9) as well as the 30% coarse
RCA mix (Fig. 10). Fig. 10 shows the 30% coarse RCA mix. It can
be seen from the gure that the RCA particles behave very similarly
to the coarse virgin aggregate. The paste produces a well consoli-
dated dense ITZ around all the coarse particles, making the con-
crete system act homogenously as a three phase system. The
coarse RCA particle is composed of residual paste and aggregate
where the two components are bound to each other and reinforce
the particle as a whole, producing a coarse aggregate very similar
to the virgin aggregate. It is also noteworthy to assess the particle
shape of the coarse RCA particles, as the processing of RCA abrades
the surface of the particle and makes it smoother. In Fig. 10, the
RCA particle is composed of cement paste, as well as ne and
found to be a suitable replacement for virgin aggregates in
order to meet the criterion of a 35 MPa C-1 exposure in accor-
dance with CSA specications. Though preferable performance
was achieved through only coarse aggregate replacement (up
to 30% by volume of coarse aggregate), full granular replace-
ment (up to 20% by volume of aggregate) had also satised
the criteria.
2. Trial mixtures maintaining a w/cm ratio of 0.4 (through the use
of a HRWR admixture), exhibited equivalent fresh properties in
terms of slump retention and measured air stability to control
mixtures, regardless of the RCA replacement level up to 30%
when the incorporated RCA is used at (or above) SSD condition.
3. Compressive strength performance of all trial mixtures remains
unaffected with either coarse or granular RCA replacement
levels up to 30% and 20%, respectively. The primarily weaker
bond of the ITZ remains the dominant factor in terms of com-
pressive strength performance [8] when a quality RCA is intro-
duced in normal strength concrete.
4. Introduction of the RCA displayed minimal impact in terms of
exural strength performance, primarily due to the controlled
quality production of the RCA. Minor detriments in perfor-
mance may be attributable to the reduced mechanical bond
experienced in the ITZ through abrasion of the softer mortar
paste.
5. Similarly, an inverse relation was observed where a decline in
splitting tensile strength occurred with increasing RCA content.
Considering the relatively weaker nature of residual paste (in
tension) to its aggregate counterpart, the higher presence of
ner particles in the granular RCA variant thus contributes to
the formation of many weaker ITZs, and in turn, reduced overall
tensile performance.
6. Sufcient air entrainment remains the dominant factor in terms
of durability performance when exposed to freezing/thawing
cycles, regardless of the RCA replacement level. Comparable
performance to virgin aggregate is exhibited with the use of
RCA if the source concrete is composed of air-entrained con-
crete and suggesting that higher percentages of air-entrained
RCA may outperform traditional mixes in colder climates in
these exposure conditions.
7. Mixtures containing RCA, where a constant w/cm ratio was
maintained, would display consistent performance within test
variation (0.015%) in terms of linear drying shrinkage.
Higher drying shrinkage occurrence was observed in trial mix-
tures containing granular RCA primarily due to the presence of
ner particulates and lower restraint.
8. Due to the porous nature of the RCA, an increase in both water
sorptivity and chloride-ion permeability were observed with
increasing RCA content, especially for granular RCA where the
ner abraded mortar particles exhibit higher water sorptivity
and lower the overall density of the paste matrix. Increasing
chloride ion permeability of the lower-density paste was easily
mitigated through the use of Ground Granulated Blast Furnace
Slag (GGBFS).
 A. Lotfy, M. Al-Fayez / Cement & Concrete Composites 61 (2015) 3643
In summary, the use of properly controlled quality RCA in pro-
duction of concrete for structural applications at severe environ-
mental exposures requires minimal considerations in mix design
to satisfy code criteria. Through such developments, it is hoped
that production of RCA-concrete for structural use may further
promote sustainable development alternatives within the
Canadian market.
References
[1] Meyer C. The greening of the concrete industry. Cement Concr Compos
2009;31:6015.
[2] West J, Tighe S, Butler L. Effect of recycled concrete aggregate properties on
mixture proportions of structural concrete. Transport Res Rec: J Transport Res
Board 2012;2290(1):10514.
[3] Leelawat T, Limbachiya R. Use of recycled concrete aggregate in high-strength
concrete. Mater Struct 2000;33:57480.
[4] Tighe S, West S, Butler L. Guidelines for the selection and use of coarse
recycled-concrete aggregates in structure concrete. Transport Res Rec: J
Transport Res Board 2013;2335:312.
[5] Garber S et al. A technology deployment plan for the use of recycled concrete
aggregates in concrete paving mixtures. National Concrete Pavement
Technology Center, Institute for Transportation, Iowa State University; 2011.
[6] DeBrito J, Rosa A, Pedro D, Matias D. Durability of concrete with recycled
coarse aggregates: inuence of superplasticizers. American Society of Civil
Engineers; 2014.
[7] West J, Tighe S, Butler L. The effect of recycled concrete aggregate properties on
the bond strength between RCA concrete and steel reinforcement. Cem Concr
Res 2011;41:103749.
[8] Otsuki N, Miyazato S, Yodsudjai M. Inuence of recycled aggregate on
interfacial transition zone, strength, chloride penetration and carbonation of
concrete. J Mater Civ Eng 2003;15:44351.
[9] Tabsh S, Abdelfatah A. Inuence of recycled coarse aggregates on strength
properties of concrete. Constr Build Mater 2009;23:11637.
[10] Hansen T. Recycled aggregates and recycled aggregate concrete: second state-
of-the-art report, developments 19451985. Mater Struct 1986;19:20146.
[11] Chern J, Hwang C, Tsai D. Research and development of HPC in Taiwan. Concr
Int 1995;17:716.
43
[12] Anderson K, Uhlmeyer J, Russell M. Use of recycled concrete aggregate in PCCP:
literature search. WSDOT research report WA-RD 726.1. Olympia, WA:
Washington State Department of Transportation (WSDOT): 37.
[13] Maleev M, Radonjanin V, Marinkovic S. Recycled concrete as aggregate for
structural concrete production. Sustainability 2010;2:120425.
[14] Sagoe-Crentsil K, Brown T, Taylor A. Performance of concrete made with
commercially produced coarse recycled concrete aggregate. Cem Concr Res
2001;31:70712.
[15] Cecen C, Ozturan T. Effect of coarse aggregate type on mechanical properties of
concretes with different strengths. Cem Concr Res 1997;27:16570.
[16] Padmini A, Ramamurthy K, Mathews M. Inuence of parent concrete on the
properties of recycled aggregate concrete. Constr Build Mater
2009;23(2):82936.
[17] Meddah M, Zitouni S, Belaabes S. Effect of content and particle size distribution
of coarse aggregate on the compressive strength of concrete. Constr Build
Mater 2010;24:50512.
[18] Etxeberria M, Vzquez E, Mar A, Barra M. Inuence of amount of recycled
coarse aggregates and production process on properties of recycled aggregate
concrete. Cem Concr Res 2007;37:73542.
[19] Nagataki S, Sacki T, Hisada M, Gokce A. Freezing and thawing resistance of air-
entrained concrete incorporating recycled coarse aggregate: the role of air
content in demolished concrete. Cem Concr Res 2004;34:799806.
[20] Coventry K, Bacon J, Richardson A. Freeze/thaw durability of concrete with
recycled demolition aggregate compared to virgin aggregate concrete. J Clean
Prod 2011;19:2727.
[21] Matias D, Brito J, Rosa A, Pedro D. Durability of concrete with recycled coarse
aggregates: inuence of super plasticizers. J Mater Civ Eng 2014;26(7).
[22] Jaturapitakkul C, Chalee W, Rattanachu P, Somna R. Effect if the water to
binder ratio and ground y ash on properties of recycled aggregate concrete. J
Mater Civ Eng 2012;24(1):1622.
[23] Berndt M. Properties of sustainable concrete containing y ash, slag, and
recycled concrete aggregate. Constr Build Mater 2009;23:260613.
[24] Padayachee N, Olorunsogo F. Performance of recycled aggregate concrete
monitored by durability indexes. Cem Concr Res 2002;32:17985.
[25] Houehanou H, Jolin M, Escaft P, Gagne R. Study of the relationship between
scaling resistance and sorptivity of concrete. Can J Civ Eng 2011;38:123848.
[26] Medina C, Zhu W, Howind T, DeRojas I. Inuence of mixes recycled aggregate
on the physicalmechanical properties of recycled concrete. J Clean Prod
2014;68:21625.